Animal models of stroke, notably transient middle cerebral artery occlusion (MCAo), are used to assess the efficacy of pharmacological and transplant treatments. Long-term studies (>1 month) of the functional effects of treatments in animal models are required to predict treatments likely to improve dysfunctions associated with stroke damage. These pre-clinical studies require (1) optimum post-operative care to ensure long-term survival, (2) methods for assignment of rats to groups with equivalent impairments to reduce variability and enhance detection of treatment effects, and (3) behavioural tests that detect long-term stable deficits. For long-term functional assessment, a battery of behavioural tests sensitive to a range of deficits observed after MCAo was developed. The bilateral asymmetry test evaluated the time course of sensory neglect. Deficits of motor integration were examined in the footfault test, and motor bias was assessed by pharmacological stimulation of rotation. The water maze was used to detect long-term deficits in spatial information processing. Long-term differences between control and MCAo animals in this battery of tests indicate that the protocol provides an efficient assessment suitable for evaluating treatment outcomes in pre-clinical studies of stroke, and that the post-operative care procedure and method of assignment to groups were effective.